利用基于推选体制（Lifting Scheme)的小波变换实现算法，构造出了与具有紧支集的双正交小波滤波器对应的小波变换和逆小波变换的算子矩阵，这些矩阵是大小与信号长度相等的可逆常数矩阵。这些矩阵为从理论的角度研究双正交小波滤波器的性能提供了一个工具。利用这些矩阵，从理论的角度对不同的双正交小波滤波器的去相关性能进行了分析比较研究，并和其他常用的正交变换，如DCT、DFT进行比较，给出了定量的比较结果。比较结果表明，在高压缩比的情况下，（9，7）滤波器具有好的性能，而在中等压缩比下，（5，3）滤波器有比较好的性能。

In this paper, based on the Lifting Scheme algorithms of the wavelet transforms with the compactly supported biorthogonal filters, the operator matrices of the forward wavelet transforms and the associated inverse wavelet transforms are constructed. These matrices are the inversive constant ones and have the size as the same as the length of the signal. These operator matrices are very convenient tools to analyze the properties of the biorthogonal wavelet filters in a theoretical way. These operator matrices can also be used for analysis the performance of the biorthogonal wavelet filters in different applications. In this paper, by these matrices, the de correlation performance of several biorthogonal wavelet filters in data compression application is studied in a view of theoretical analysis. The qualified quantities are given in comparison with other popular orthogonal transforms such as DCT and DFT. Experimental results demonstrate that different wavelet filters have different performance in the data compression applications. In the case of the high compression ratio, (9,7) filter has the better performance while in the case of the moderate compression ratio, (5,3) filter has the better performance.